RESUMEN
INTRODUCTION: The "cognitive reserve" hypothesis states that education provides a better tolerance to brain pathological processes. Consequently, the development of dementia would require more cognitive loss in more educated people. We examine in a cohort of patients with different levels of cognitive impairment, the relationship between the amount of cognitive loss and educational attainment. METHODS: A total of consecutive outpatients were assessed prospectively with an extensive clinical and neuropsychological protocol, followed up for 6 months, diagnosed by a central expert committee as demented (DSM-III-R criteria) or cognitively impaired. They were included in the Spanish Multicentric Registry of Incident Cases of Dementia. Premorbid verbal fluency (pVF), as a marker of previous cognitive function, was estimated by a validated predictive model. Cognitive loss was defined as the difference between estimated pVF and present observed verbal fluency (oVF) in two ways: as an absolute Z-score (pVF-oVF/sigma of residuals in the normative sample) and as a relative quotient-score (pVF-oVF/pVF). RESULTS: A two-way ANOVA showed that clinical stage (Z-scores: F=75.88; p<0.0001; quotient-scores: F=126.56; p<0.0001) and also education (Z-scores: F = 83.51; p=0.0001; quotient-scores: F=34.88; p=0.0001) were positively related with both estimates of cognitive loss. CONCLUSIONS: Our data confirm that education, or some other related factor, determines the amount of cognitive loss needed for clinical expression of the pathological illness and gives the individuals more tolerance to these processes.